EP0759313A2

EP0759313A2 - Device for varying the threshold detection level of a sensor

Info

Publication number: EP0759313A2
Application number: EP96850125A
Authority: EP
Inventors: Jan Ljungström
Original assignee: Pacesetter AB
Current assignee: Pacesetter AB
Priority date: 1995-07-04
Filing date: 1996-06-27
Publication date: 1997-02-26
Also published as: SE9502430D0; JPH0919502A; US5669392A; EP0759313A3

Abstract

This invention relates to a device capable of automatically adjusting the threshold detection level of a detector unit for detecting physiological events, e.g. heart contractions, comprising:
sensor means (2, 10) for sensing a cardiac signal, sampling (5) and memory (6) for collecting and storing samples of said cardiac signal,
whereby it further comprises:
timer (7) for starting the sampling means (5) and stopping the sampling means (5) after a predetermined time, analysis means (8) for the analysis of the stored cardiac signal to generate a first set of values which represent physiological events, such as mean peak signal value, the average maximum signal slope or an integrated signal value, analysis means (8) for calculating at least one derived parameter from the whole of said first set of values and analysis means for automatically varying the threshold detection level of the sensor means (2, 10) as a function of said at least one derived parameter.

Description

The invention relates to a device for adjusting the threshold detection level of a sensor, particularly a sensor for sensing physiological signals for use with an implantable pacemaker or cardiac diagnostic device, to take into account the level of the sensed signal.

Background of the invention.

Sensors are used medically to detect signals which relate to the physiological activity of a subject and which can be used to aid diagnosis or treatment. Various different parameters can be measured, for example, pulse rate, respiration rate, blood pressure, temperature, etc. The choice of which parameter or parameters to detect is decided by a physician and is adapted to the specific circumstances of each subject.
Sensors can be used to detect the cardiac activity of pacemaker users by measuring the electrical signals produced by the heart when it beats. Each heart contraction is preceded by an electrical pulse and the detection of these electrical pulses is used to control pacemakers. To avoid noise and signals generated by non-cardiac muscle activity being erroneously sensed as a cardiac event it is usual to set a sensor threshold detection level below which all signals are ignored. The choice of threshold detection level is of critical importance. Setting the threshold detection level too high can result in "undersensing"- a condition where cardiac events are not sensed which could mean that the pacemaker loses synchronisation with the intended natural cardiac rhythm or delivers extra pacing stimuli at inappropriate times, for example, just after a successful capture which was not sensed. Setting the threshold detection level too low will result in non-cardiac electrical activity erroneously being sensed as cardiac events. This is known as "oversensing" and can lead to the pacemaker delivering stimulation pulses at inappropriate times, can cause high pacing rates and can lead to pacemaker-mediated tachycardia. Hence it is important that a correct threshold detection level is set.
Unfortunately the impedance of the signal path in the patient's bddy varies considerably as time passes. This change in impedance affects the amplitude of the signal and the sensor sensitivity has to be adjusted to take account of this change. Long term changes, over a period of days, weeks or months can be caused by lead maturation or chronic changes in cardiac activity, while short term changes occur between heart beats and can be caused by changes in the patient's activity or changes in the chest shape due to respiration.

Prior Art

A method for adjusting the sensor sensitivity is shown in prior art document US-A-4 708 144 which describes a method of calculating the average peak amplitude of a cardiac signal over a period of minutes and subsequently varying the amplification of the cardiac signal in accordance with the average peak amplitude. As the average peak amplitude is calculated over a period of minutes, this method reacts slowly to changes in the peak amplitude and cannot take into account beat-to-beat variations in cardiac activity.
EP-A-580 894 describes a method for adjusting the threshold of a sensor when the running average value over, for example, 18 heart beats, of the margin by which a measured signal exceeds a lower threshold falls outside upper and lower boundaries. Once this occurs a new threshold is established with new upper and lower boundaries. The requirement that the average value must first fall outside the region between the upper and lower boundaries means that the reaction to changes in the signal takes place after a delay and in a step like fashion. This method cannot take into account beat-to-beat variations in cardiac activity. Furthermore, the use of a running average means that only a small amount of the total information in the signal is stored and hence is available for analysis and hence more sophisticated analysis of the stored signal is not possible.
An object of the present invention is to provide a device for automatically adjusting the sensitivity of a sensing device for an implantable pacemaker in which the beat-to-beat variations in sensed cardiac activity are taken into account.

Summary of the invention

The sensing device is controlled by circuitry which stores samples representing a small number of cardiac events and analyses the stored signal to calculate a new threshold detection level.

Brief description of the figures

Figure 1 is a diagram representing a threshold adjusting device in accordance with one embodiment of the present invention.
Figure 2 is a flow diagram according to one embodiment of the present invention.

Detailed description of the invention

A first embodiment of a sensing device for a pacemaker according to the invention, contains circuitry for sampling an input signal, timer means for starting the sampling and stopping the sampling after a predetermined time, a memory for storing said input signal samples, an analysing means for analysing said input signal samples to generate a first set of values representing cardiac events, calculating means for calculating a threshold detection level and means for producing a output when said input signal fulfils certain conditions. It also contains communication means by which it can be programmed by a physician or technician and by means of which it can pass information to a physician or technician.
The sensing device is connectable to a lead which picks up electrical signals from the heart. The signal is sampled at regular intervals in order to detect cardiac events. The analysing circuitry analyses the signal to determine when a cardiac event has probably taken place. One way of determining if a cardiac event has taken place is to set various conditions which must be fulfilled before a signal is accepted as representing a cardiac event. An example of such conditions would be that after a cardiac event has been determined to have taken place, the signal must first go to below zero, then reach a peak which is above the sensor threshold and finally drop to zero again. Once these conditions have been achieved it is assumed that a cardiac event has taken place and this information is used by the pacemaker to control its pacing.
In the first embodiment of the invention the sensor threshold can automatically be adjusted in response to variations in the peak value of the input signal. This is achieved by storing the samples of the input signal and analysing them. In order to reduce cost the samples can be stored in a small memory which can store only a limited number of samples and a limited number of signal levels. Typically, samples could be taken every 4 mS for 4 seconds and the signal resolution limited to an 8 bit value for each sample. The timing means controls the storing of the input signal and can be programmed to continuously store the most recent samples or can be programmed to only store samples at certain intervals e.g. once an hour, once a day, once every 36 hours, once a week, once a month, etc. Once the samples have been stored the analysis means analyses the stored signal values using, for example, the above mentioned conditions to find out how many cardiac events took place in the 4 seconds and to determine the peak signal level associated with each cardiac event. Each of these peak signal levels are stored and together they form a set of first representative values representing cardiac events. A derived parameter which can be used to represent the stored signal is then derived from the set of first representative values. In this first embodiment the derived parameter is the mean value of said set of first representative values. Other derived parameters could be the maximum slope of the curve representing a cardiac event, or an integrated value such as the area under the curve, or any other such parameter. The threshold detection level is then set at a level which is a predetermined or programmable function of said derived parameter, or the level can be found from a look-up table.
Figure 1 shows a diagram representing a threshold adjusting device in accordance with one embodiment of the present invention. The device shown can be built into a pacemaker or made as a separate unit. Heart 1 is connected by electrode 2 to threshold adjusting device 3. Electrode 2 carries a signal 4 which represents cardiac activity. Signal 4 shows in figure 1 how the amplitude of the electrical signal from the heart varies with respect to time. Signal 4 shows four QRS signals which have different maximum values, x1, x2, x3, x4. Signal 4 is sampled by sampling means 5, which could be an A/D converter, and the digital signal stored in memory 6 at an interval controlled by timer 7. The stored signal is analysed, in a manner described later, by analysis means 8, which can be a microprocessor or similar device, to determine a new detector threshold value 9. Memory 6 comprises RAM and can be internal or external to the microprocessor 8. Memory 6 contains the operating commands for instructing the microprocessor how to perform the analysis of the cardiac signal and subsequently control the variation in the sensor threshold detection level. Memory 6 stores the detector threshold values 9 calculated by the microprocessor 8. It can also contain a look-up table of sensor threshold detection levels. Detector threshold value 9 is communicated to threshold detector 10 which filters the digital signal from A/D converter 5 before sending the signal to therapy circuits 11. Therapy circuits 11 control the functioning of a pacemaker 12.
In the embodiment shown in figure 1 the threshold adjusting device 3 has communication circuit connected to a telemetry coil 14 for sending signal to, and receiving signals from, a suitably equipped operator. This allows the device to be reprogrammed and also to transmit stored information for further analysis as necessary.
Figure 2 shows a flow diagram for a microprocessor based circuit for adjusting a sensor threshold detection level every 24 hours.
In this embodiment microprocessor 8 has counter S which is incremented by one every second at stage 21. Counter S records the time S which has elapsed since the last time the threshold detection level was adjusted. Counter S is monitored by microprocessor 8 at stage 22 and when counter S reaches 86 400, which means 24 hours have elapsed, microprocessor 8 sets timer T to zero and counter N to zero at stage 23. At stage 24 the cardiac signal is sampled and its amplitude stored in memory M in space number N. The microprocessor 8 waits 4 mS at stage 25 and then increments both timer T and counter N by one at stage 26. Stage 27 checks how many samples have been made. If counter N is less than 1000 then microprocessor 8 returns to stage 24 and makes another sample. If counter N has reached 1000 then the sampling is stopped and the stored signal is analysed at stage 28 to identify cardiac events as explained previously. The peak signal value (x1, x2, x3, x4...xn) is calculated for each event at stage 29 and a first derived parameter, the mean peak signal value is calculated at stage 30. Microprocessor 8 finds the new sensor detector threshold at stage 31 by looking up in a look-up table the sensor detector threshold level which corresponds to the calculated mean peak signal value and the threshold detector level is set to the new level. The timer S is reset to zero at stage 32 and the cycle starts again from stage 21.
Instead of, or in addition to, the peak signal value (xn) some other value such as maximum positive or negative slope, could be determined from the stored signal value and used as a second representative value to represent the occurrence of a cardiac event. A second derived parameter representing, for example, the average value of these second representative values for the stored cardiac events could be calculated and the threshold detection level then set at a level which is a predetermined or programmable function of the second derived variable or the level can be found from a look-up table.
In another embodiment of the invention representative values are compared against the average value and the representative value which differs most from the average value is ignored and a new average value calculated. The threshold detection level is then set at a level which is a predetermined or programmable proportion of the latter average value or the level can be found from a look-up table.
In a further embodiment of the invention the representative value which is highest (e.g. x3 in figure 1) is ignored when the average value is calculated. As before, the threshold detection level is set at a level which is a predetermined or programmable proportion of the average value or the level can be found from a look-up table.
It is of course possible to calculate more derived parameters and to use a combination of two or more of them to establish the new threshold detection level.
It is not intended to restrict the invention to the above mentioned embodiments but to include all embodiments and modifications which fall under the scope of the claims.

Claims

A medical diagnostic device comprising:
a) sensor means (2, 10), for sensing a periodic physiological signal, connected to,

b) sampling means (5), for collecting samples of said periodic physiological signal, coupled to,

c) memory (6), for storing said samples of said periodic physiological signal,
characterized in that it further comprises
d) timer (7), for starting sampling means (5) and stopping sampling means(5) after a predetermined time,

e) analysis means (8) for the analysis of the stored periodic physiological signal to identify physiological events to generate a first set of values corresponding to physiological events,

f) analysis means (8) for calculating at least one derived parameter from the said first set of values ,

g) analysis means (8) for automatically varying a threshold detection level (9) for the sensor means (2, 10) as a function of said at least one derived parameter.
A medical diagnostic device according to claim 1 characterized in that said derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
A medical diagnostic device according to any of the preceding claims characterized in that it comprises:
discriminating means for ignoring at least one value of said first set of values to form a second set of values,

means for calculating at least one auxiliary derived parameter from said second set of values, and,

means for automatically varying a threshold detection level for the sensor means (2, 10) as a function of said at least one auxiliary derived parameter and/or said derived parameter.
A medical diagnostic device according to claim 3 characterized in that the auxiliary derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
A medical diagnostic device according to claim 3 characterized in that said ignored value of said first set of values is the maximum value and/or the minimum value and/or the value which is furthest from the average value of said first set of values.
Method for adjusting the threshold detection level of a sensor means (2, 10) connectable to a medical diagnostic device wherein the method comprises the steps of;
a) sensing a periodic cardiac signal,

b) collecting samples of said periodic cardiac signal,

c) storing the samples of said periodic cardiac signal,

d) starting the sampling means and stopping the sampling means after a predetermined time,
characterized in that it further comprises
e) analysis of the stored periodic cardiac signal to generate a first set of values corresponding to cardiac events,

f) calculating at least one derived parameter from the whole of said first set of values ,

g) automatically varying a threshold detection level for the sensor means (2, 10) as a function of said at least one derived parameter.
A method according to claim 6 characterized in that derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
Method according to claim 6 characterized in that it comprises the steps of ignoring at least one value of said first set of values to form a second set of values, calculating at least one auxiliary derived parameter from said second set of values, and, for automatically varying a threshold detection level as a function of said at least one auxiliary derived parameter and/or said derived parameter.
Method according to claim 8 characterized in that auxiliary derived parameters can be one or more of: the mean peak signal value, the average maximum signal slope and an integrated signal value.
Method according to claim 6 or claim 8 characterized in that said ignored value of said first set of values is the maximum value and/or the minimum value and/or the value which is furthest from the average value of said first set of values.
A medical diagnostic device comprising:
a) sensor means (2, 10), for sensing a cardiac signal, connected to,

b) sampling means (5), for collecting samples of said cardiac signal, coupled to,

c) storage means (6), for storing said samples of said cardiac signal, characterized in that it further comprises

d) timer means (7), for starting sampling means (5) and stopping sampling means (5) after a predetermined time,

e) analysis means (8) for the analysis of the stored cardiac signal to identify cardiac events to generate a first set of values corresponding to cardiac events,

f) analysis means (8) for calculating at least one derived parameter from the said first set of values ,

g) analysis means (8) for automatically varying a threshold detection level for the sensor means (2, 10) as a function of said at least one derived parameter.
A medical diagnostic device according to claim 11 characterized in that said derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
A medical diagnostic device according to claim 11 or 12 characterized in that it comprises:
discriminating means for ignoring at least one value of said first set of values to form a second set of values,

means for calculating at least one auxiliary derived parameter from said second set of values, and,

means for automatically varying a threshold detection level for the sensor means (2, 10) as a function of said at least one auxiliary derived parameter and/or said derived parameter.
A medical diagnostic device according to claim 13 characterized in that the auxiliary derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
A medical diagnostic device according to claim 13 characterized in that said ignored value of said first set of values is the maximum value and/or the minimum value and/or the value which is furthest from the average value of said first set of values.
Method for adjusting the threshold detection level of a sensor means (2, 10) connectable to a medical diagnostic device wherein the method comprises the steps of;
a) sensing a cardiac signal,

b) collecting samples of said cardiac signal,

c) storing the samples of said cardiac signal,

d) starting the sampling means and stopping the sampling means after a predetermined time,
characterized in that it further comprises
e) analysis of the stored cardiac signal to generate a first set of values corresponding to cardiac events,

f) calculating at least one derived parameter from the whole of said first set of values ,

g) automatically varying a threshold detection level for the sensor means (2, 10) as a function of said at least one derived parameter.
A method according to claim 16 characterized in that derived parameters can be one or more of:
the mean peak signal value, the average maximum signal slope and an integrated signal value.
Method according to claim 15 characterized in that it comprises the steps of ignoring at least one value of said first set of values to form a second set of values, calculating at least one auxiliary derived parameter from said second set of values, and, for automatically varying a threshold detection level as a function of said at least one auxiliary derived parameter and/or said derived parameter.
Method according to claim 18 characterized in that auxiliary derived parameters can be one or more of: the mean peak signal value, the average maximum signal slope and an integrated signal value.
Method according to claim 16 or 18 characterized in that said ignored value of said first set of values is the maximum value and/or the minimum value and/or the value which is furthest from the average value of said first set of values.